Brucella spp., a gram-negative facultative intracellular bacterium, induces chronic infectious disease and is a potential pathogen for bioterrorism. B. abortus, B. melitensis, and B. suis are pathogenic to humans. In contrast, human infections of B. ovis and B. neotomae have not been reported, while B. canis human infections are rare. Although limited genome diversity exists among Brucella spp., these species exhibit host preference and considerable differences in virulence. However, little is known regarding the Brucella genes that contribute to intracellular survival and virulence. Therefore, understanding differences among Brucella species genomes would provide valuable clues to different pathogenic mechanisms of this species. Pathogens can be distinguished from their avirulent counterparts by specific genes or gene clusters in the genome helping to define bacterial pathogenicity. Recently, we developed a microarray of the complete B. melitensis genome sequence, the species highly pathogenic to humans. Hybridization of DNA from other Brucella species to this microarray revealed clusters (islands) in the 16M genome that were missing in the non-virulent species. Included in these islands are ORFs encoding hypothetical proteins, transporters, transposases, and transcriptional regulators that most likely aid successful establishment of Brucella in human infections. The genomic context of these islands suggests horizontal acquisition and variability in virulence and host preference among Brucella species. Now, we will create strategies to determine the importance of these newly identified Brucella genomic islands to bacterial pathogenicity. We hypothesize that genetic information contained within certain of these islands contributes to establishing an intracellular niche and persistence of Brucella within animals. Our long-term goal is to characterize at the molecular level the role of Brucella genes within these genomic islands for their contribution to Brucella pathogenicity and survival. Defining the contributions of genomic islands to Brucella pathogenesis will aid in developing future Brucella vaccines. The following specific aims are proposed: Aim 1. We will determine the contribution of genomic islands to Brucella pathogenicity. (a) We will delete genomic islands from virulent B. melitensis. (b) We will determine if genomic islands have functional importance for Brucella survival in macrophages. Aim 2. We will determine persistence and dissemination of mutants in mice. We will compare the in vivo dissemination and persistence of genomic island mutants to virulent B. melitensis using biophotonic imaging in mice. Aim 3. We will determine gene transcription of B. melitensis and GI mutants in macrophages as well as gene transcription by mammalian cells responding to infection. (a) We will determine transcriptional differences between virulent B. melitensis and GI mutants in macrophages. (b) We will identify differences in macrophage transcription following infection of mutant and virulent Brucella.